The long-term goals of this work are to understand the molecular signaling pathways involved in the development and regeneration of inner ear sensory and neuronal populations. The studies done during the initial grant period were directed at discovering the upstream genes that regulate expression of the growth factor Bone Morphogenetic Protein 4 (BMP4) and its role in the formation of sensory hair cells (HC) in inner ear development. The work done to date has demonstrated that BMP4/antagonist cascades are critical for normal inner ear morphogenesis (Gerlach et al, 2000;Gerlach-Bank et al, 2002 and Gerlach-Bank et al, 2004) and has identified a novel BMP4 promoter, which is expressed in the inner ear and which is repressed by retinoic acid (RA), explaining why RA-treated inner ears resemble BMP4 antagonist-treated ears (Thompson et al, 2003). In this renewal application, attention is focused on the downstream genes, particularly the role of ZIC genes in inner ear development. The hypothesis that directs this work is that ZIC genes are critical transcription factors that serve as regulatory genes, channeling undifferentiated otocyst precursor cells either to a sensory neuron or sensory hair cell (HC)/supporting cell (SC) fate from a common precursor cell in the otocyst. This laboratory has shown that both ZIC1 and ZIC 2 are expressed at the right time and in the right cells to play critical regulatory roles in this cell lineage selection (Warner et al, 2003). In addition, studies of immortalized inner ear cell lines and ZIC2 knockdown mouse models in this laboratory (Gerlach-Bank et al, in prep.) have provided additional support for this hypothesis. The experiments in this proposal therefore investigate the role of ZIC genes and the atonal class downstream genes they control in sensory cell and sensory neuron lineage specification in the inner ear. For these experiments immortalized otocyst cell lines that represent neuron-like, hair cell- (HC) or supporting cell (SC)-like, and neuronal/sensory precursor-like IMO cell populations (precursor-like) will be used (Germiller et al, submitted). Findings from the cell line experiments will then be tested in functional experiments in "real" inner ears in embryos of chick and ZIC knockout and knockdown mice. The hypotheses to be tested include: ZIC1 genes control sensory neuron formation through regulation of Neurogeninl and NeuroD and ZIC2 genes control sensory hair cell formation in the inner ear by regulating Math1. ZIC1 is also known to downregulate MathI (Ebert et al, 2003) and in some cases in the CNS, BMP4 is thought to upregulate Mathl (Ebert et al, 2003;Alder et al, 1999) but whether this is through effects on ZIC genes is unknown. Gene profiling through microarrays and analyses of loss- and gain-of-function experiments in these model systems will allow tests of all these hypotheses.